That's an extra $100 in what is essentially a solar tax, not an extra $100 vs not having solar at all. IOW, his bill will go to about $100 per month vs the $3 per month he was paying before, which was with him providing power back to the grid, not taking. Considering all the complaining about the grid needing to be updated and how we're thinking about paying trillions more in taxes for infrastructure, you'd think we'd want more people generating their own power, not less. But then that assumes your goal isn't to support and maintain the local electric monopoly.
This will continue to be a problem as more people move to solar. The existing bureaucracies and government regulators will fight tooth and nail to maintain the status quo, which means maintaining grid and production these customers may not even use, and if they do, will use less over time. The solution for homeowners of course is to go off-grid and tell the electric companies to pound sand, which I'm sure many will, although eventually governments will likely try to impose a general 'electric grid' tax on all citizens whether they use grid-connected power or not, all in an effort to maintain current cash flows. After all, when 90% of your customers leave to generate their own power, how much would the remaining 10% have to pay to keep the system in place? They won't be paying 10 times more that's for sure - they wouldn't stand for it - so you know where that money will be coming from.
Residential solar grid-tie subsidies and the utility customer death spiral are major concerns in some circles.
You are absolutely right that as solar gets cheaper, and to the extent that batteries are affordable, those with means will go off-grid and ditch their utility connections. This means fewer houses per mile of electrical wire and fewer KWh's per hour of maintenance labour. That means more expensive power, which will drive more people to go off-grid.
Charging people to not use utility power only goes so far, especially since it's the wealthiest who will change first.
The reason this is related to residential grid-tie subsidies, where the utility buys power from residential solar installations are more than wholesale power rates, is that few people take the time to figure out how much of the monthly electrical bill is for the power, how much for the delivery, and how much for reliability.
Most of the time, wholesale electricity rates are one single digit cents per KWh. With solar power the wholesale rates often go negative during the early afternoon, only to shoot up just after the solar passes its peak production. People expecting to get paid residential rates (ie. net metering) with grid-tie systems are either going to be disappointed, or subsidized by people without solar systems. The wholesale rates are simply too low.
Then there's delivery, really the most expensive part. Maintaining all those hundreds of thousands of miles of high power wiring and transformers is expensive. For grid-tie solar to be sustainable, the wiring to homes needs to be upgraded and that use paid for. I fully expect the day when grid-tie systems must pay a delivery charge per-KWh supplied to the grid.
In a grid-tie system, the homeowner is a part-time energy supplier and in the long run the accounting needs to match that reality. Anything else either requires never-ending tax payer money, or will cause utilities to become unviable as businesses. That'd be very bad when the next high-demand, low-supply week comes along. Reliability in the face of just about any scenario costs real money.